By using sonar imaging, this paper presents a new algorithm for the clustering of seabed types based on the self-organizing feature maps （SOFM） neural network. The theory as well as data processing is studied in detail. Some valuable conclusions and suggestions are given 相似文献
A model integrating geo-information and self-organizing map (SOM) for exploring the database of soil environmental surveys was established. The dataset of 5 heavy metals (As, Cd, Cr, Hg, and Pb) was built by the regular grid sampling in Hechi, Guangxi Zhuang Autonomous Region in southern China. Auxiliary datasets were collected throughout the study area to help interpret the potential causes of pollution. The main findings are as follows: (1) Soil samples of 5 elements exhibited strong variation and high skewness. High pollution risk existed in the case study area, especially Hg and Cd. (2) As and Pb had a similar topo-logical distribution pattern, meaning they behaved similarly in the soil environment. Cr had behaviours in soil different from those of the other 4 elements. (3) From the U-matrix of SOM networks, 3 levels of SEQ were identified, and 11 high risk areas of soil heavy metal-contaminated were found throughout the study area, which were basically near rivers, factories, and ore zones. (4) The variations of contamination index (CI) followed the trend of construction land (1.353) > forestland (1.267) > cropland (1.175) > grassland (1.056), which suggest that decision makers should focus more on the problem of soil pollution surrounding industrial and mining enterprises and farmland.
The variability of bottom dissolved oxygen (DO) in Long Island Sound, New York, is examined using water quality monitoring data collected by the Connecticut Department of Environmental Protection from 1995 to 2004. Self-organizing map analysis indicates that hypoxia always occurs in the Narrows during summer and less frequently in the Western and the Central Basins. The primary factor controlling the bottom DO, changes spatially and temporally. For non-summer seasons, the levels of bottom DO are strongly associated with water temperature, which means DO availability is primarily driven by solubility. During summer, stratification intensifies under weak wind conditions and bottom DO starts to decrease and deviate from the saturation level except for stations in the Eastern Basin. For the westernmost and shallow (<15 m) stations, bottom DO is correlated with the density stratification (represented by difference between surface and bottom density). In contrast, at deep stations (>20 m), the relationship between oxygen depletion and stratification is not significant. For stations located west of the Central Basin, bottom DO continues to decrease during summer until it reaches its minimum when bottom temperature is around 19–20 °C. In most cases the recovery to saturation levels at the beginning of fall is fast, but not necessarily associated with increased wind mixing. Therefore, we propose that the DO recovery may be a manifestation of either the reduced microbial activity combined with the depletion of organic matter or horizontal exchange. Hypoxic volume is weakly correlated to the summer wind speed, spring total nitrogen, spring chlorophyll a, and maximum river discharge. When all variables are combined in a multiple regression, the coefficient of determination (r2) is 0.92. Surprisingly, the weakest variable is the total nitrogen, because when it is excluded the coefficient r2 only drops to 0.84. Spring bloom seems to be an important source of organic carbon pool and biological uptake of oxygen plays a more crucial role in the seasonal evolution of bottom DO than previously thought. Our results indicate that the reassessment phase of the Long Island Sound Total Maximum Daily Load policy on nitrogen loading will most likely fail, because it ignores the contributions of the spring organic carbon pool and river discharge. Also, it is questionable whether the goal of 58.5% anthropogenic nitrogen load reduction is enough. 相似文献
